CompoundSemi Online - The Original Compound Semiconductor Industry Newspaper

April 21, 2005...We talk a lot about the packaging of compound semi devices, but in this communications-driven international marketplace, how a company's corporate message gets packaged can make all the difference in the world in actually scoring new business. As compound semi device manufacturers look for the next "Oh Gosh/Gee Whiz" trend to follow, what we're seeing is an astonishingly fast race to offer GaN electronic device foundry capability. At latest count by one of the major potential customers for upgrades in communication device testing equipment, over a dozen companies were already offering or saying they would offer GaN-based electronic device foundry products and services in the near future. The person looking for and characterizing samples found only two companies - one in Europe and one in the USA - interested in actually engaging with a potential customer for their foundry services, present or upcoming. Evidently most of the others were actually offering "smoke and mirrors."

I was astonished there were so many names on the list. The caller was asking if he'd left any out, so we exchanged notes, my list being much shorter. At last compilation, here were all the names we could come up with. These are not... and I repeat, NOT, in any sort of order or ranking, whatsoever: Northrop Grumman, Raytheon, Rockwell, HRL, RFMD, Cree, QinetiQ, MicroGaN, Nitronex, Fujitsu (and possibly other Japanese companies, SEI perhaps), TriQuint, IMEC, SET Inc., and Osemi. What I'd like to hear from our readers, are any additions, subtractions, or qualifiers regarding the list.

I personally haven't "run" this list, but our staff has been chartered to see what they can come up with. Your input would be greatly appreciated. There is some excellent work being done in universities, of course, and the trend these days is to spin off a commercial venture from a university, hoist a corporate flag, and see if anyone salutes it. SET Inc. has done that, as a spinoff of the University of South Carolina. The WBG operation is being run by Asif Khan who truly did pioneer GaN electronics. Nitres had the same sort of scenario, spinning off of UC Santa Barbara, which eventually sold to Cree. Nitres is now Cree's Santa Barbara research center, which is doing true state of the art R&D.

One of the companies that got especially good marks was a German foundry called MicroGaN. Their website doesn't provide much background about the company, but it does a very good, no-nonsense job of explaining the technology basics and product/capability offerings. I especially liked their compact description of what GaN-based electronics is all about and how to use it. It's a good basic primer to those new to the field. Here's how they explain it (edited only slightly and compacted):

"The wide band gap material GaN (Gallium-Nitride) and its ternary compounds AlGaN (Aluminium-Gallium-Nitride) and InGaN (Indium-Gallium-Nitride) are characterized by high breakdown voltages. This enables employment of high operation voltages. Therefore GaN utilizes power transistors with power densities and high output impedance above the classical GaAs/InP “High Electron Mobility Transistors“ (HEMTs) and “Hetero-junction Bipolar Transistors“ (HBTs). The major advantages of the GaN transistor technology can be summarized as follows: 1) high operation voltage, 2) high output impedance, 3) high power density, and 4) high temperature durability. These device-properties effect a reduction of circuit complexity, raise efficiency and bandwidth, increase lifetime of modules and last but not least reduce cost of systems."

We'll be getting in touch with MicroGaN's principals and will provide more information ASAP in our news pages. Located in Ulm, Germany, the company apparently was founded in 2002 by Ingo Daumiller and Mike Kunze. According to their website, their foundry offerings include three device-configurations: InGaN-HEMTs, AlGaN-HEMTs, and GaN-MESFETs on two substrate types: sapphire and silicon.

MicroGaN's offerings aside, I informed my big company evaluator that the field is still in its nascent stage, despite what people say and do. One of the furthest along with GaN on Si, is Nitronex having proved themselves on TriQuint's production line. And I told him all three companies acting as the primes for the new DARPA WBG initiative ( Ref: our March 14th news and editorial coverage and April 18th news). All are definitely in line to become major GaN electronics foundries, especially TriQuint, which is more likely to actually open their doors in a pure play manner than either Northrop Grumman or Raytheon. But it's still early in the game to expect much real product. I also gave him the names of some good resource people who know their GaN.

What I found most disturbing in my conversation with the GaN market evaluator was the subtle deception he was seeing, feeling, and believing about what a company actually had to offer. Veterans of the silicon wars in the 1970s and early compound semi "burn" victims of the 1980s would call that classic "smoke and mirror" marketing, thus the inspiration for the title of this editorial. Would it hurt to be upfront and open about what you actually have to offer versus what you'd like to offer in the future? It would certainly be refreshing.

Packaging smoke and mirrors takes a lot of bubble wrap. Smoke is hard to contain. Mirrors are fragile. It's much easier to package reliable GaN-based IC devices that meet the specs, and ship those as samples. If you have them. If you don't, say so and let the customer get on with the business at hand. If you stay upfront with them, they'll come back and sample your goods when the goods are really ready. If you deceive them now, the customer may never trust you again.

Sony & Toshiba Working on Hybrid Blue Laser DVD Solution
Jo Ann McDonald, Founding Editor

JDS Uniphase Tightens Belt Further With Plant Consolidations
CompoundSemi News Staff

Sony Open to DVD Format Discussion
CompoundSemi News Staff

Edwards Technologies Ltd. Wins Golden Peak Award in Taiwan
CompoundSemi News Staff

Goldeneye Appears on Radar With New Patent
Scott McMahan

Raytheon and Cree Confirm $26.9 Million DARPA Contract for GaN RF Components
CompoundSemi News Staff

Packaging Smoke and Mirrors

Sony & Toshiba Working on Hybrid Blue Laser DVD SolutionJo Ann McDonald, Founding Editor

JDS Uniphase Tightens Belt Further With Plant ConsolidationsCompoundSemi News Staff

Sony Open to DVD Format DiscussionCompoundSemi News Staff

Edwards Technologies Ltd. Wins Golden Peak Award in TaiwanCompoundSemi News Staff

Goldeneye Appears on Radar With New PatentScott McMahan